一种快速有效的指纹识别算法

A fast and efficient fingerprint recognition algorithm

指纹图像是由交错排列的脊线和谷线组成,包含了丰富的纹理信息。为了获得更为有效的纹理信息,首先将图像分解为3个级别的区域块,来校正小尺寸块的方向角,以确定指纹的中心点位置,进而提高算法的运算速度;然后利用小波变换将指纹图像由空间域转换到频率域,在频率域进行多尺度分析,提取指纹图像在不同频率和方向上的局部纹理信息,提高指纹识别的准确率。该算法在以ARM11为处理器的嵌入式系统中实现。实验结果表明,该算法能够快速有效地识别指纹图像。

The fingerprint image is composed of staggered ridges and valleys,uhich contains rich texture information. In order to obtain more effective texture information,firstly,the image was decomposed into 3 levels of block to correct the direction angle of the small size block. Thus the center point position of the fingerprint was determined and the operation speed of the algorithm was improved. Then the fingerprint image was converted from the spatial domain to the frequency domain by wavelet transform. The multi-scale analysis was used to extract local texture information of the fingerprint image in different frequencies and directions. So the accuracy rate of fingerprint identification was improved. The algorithm is implemented in the embedded system of ARM11. The experimental results show that the proposed algorithm could quickly and effectively identify the fingerprint image.